Data Flow Controller for Reconfigurable Computers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reconfigurable computers face limitations in programming flexibility and efficiency due to inflexible data flow control logic and inefficient dynamic switching methods, which hinder their ability to handle applications with multiple data sources and require complex, costly interconnection networks.
Innovation Solution
A novel data flow controller that allows for configurable operation conditions, enabling elements to perform operations based on various availability conditions, and a dynamic switch element dedicated to flow control, which reduces complexity and overhead in the interconnection network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional data flow control logic is used (operation only when all selected input sources are available), then control simplicity is maintained, but programming flexibility and applicability to diverse data sources is limited
Solution Approach 1:
The patent implements dynamic data flow control logic that can adapt its operation mode based on configuration settings. The control logic dynamically switches between different operational modes (e.g., waiting for all inputs vs. operating on first available input) to match the specific application requirements, thereby achieving programming flexibility without permanently increasing hardware complexity
Solution Approach 2:
The patent changes the operational parameters of the data flow control logic by introducing configurable modes that alter the availability conditions for operation. By parameterizing the control behavior (e.g., through configuration bits or control signals), the system achieves versatility across different applications while maintaining a relatively simple underlying hardware structure
2Adaptability or versatility
If a large, dynamically switched interconnection network is implemented, then dynamic switching capability is achieved, but chip size, power consumption, and implementation cost increase significantly
Solution Approach 1:
The patent segments the interconnection network into smaller, manageable switching elements or stages. Instead of implementing a single large dynamic switch, the network is divided into multiple smaller switching units that can be independently controlled, reducing the complexity and resource requirements of each individual element while maintaining overall dynamic switching capability
Solution Approach 2:
The patent merges switching functionality with existing functional elements in the reconfigurable computer architecture. By integrating switch control into the fabric of the existing architecture rather than adding a separate comprehensive switching layer, the system achieves dynamic switching capability while minimizing additional chip size, power consumption, and implementation complexity
3Productivity
If functional elements are used for both primary operations and dynamic switching, then hardware utilization is maximized, but switching overhead increases and functional elements become unavailable for their primary function
Solution Approach 1:
The patent extracts the dynamic switching function from the primary functional elements and implements it through dedicated control logic or separate switching mechanisms. This allows functional elements to focus on their primary computational operations while switching is handled by specialized control structures, reducing switching overhead and preventing functional elements from becoming unavailable for their main purpose
Data Source
AI summary
A data flow controller for reconfigurable computers. The novel data flow controller includes a first circuit for selecting one of a plurality of operating conditions and a second circuit for determining if the selected condition is met and outputting a control signal accordingly. In an illustrative embodiment, the operating conditions include: when all enabled data available signals are asserted and all enabled space available signals are asserted; when any enabled data available signal is asserted and all enabled space available signals are asserted; when all enabled data available signals are asserted and any enabled space available signal is asserted; and when any enabled data available signal is asserted and any enabled space available signal is asserted. By allowing a configurable element to operate under different possible conditions, data flow signals can also then be used to control what operation the element performs, in addition to controlling when.